Machine Learning and Data Mining – MLDM

Course Coordinator: Fotios Kokkoras,      ECTS: 7.5,      Semester: A (C)

Syllabus

  • Machine Learning (what it is, why we care, examples of problems, historical review, algorithm categorization).
  • Machine Learning System Design, Machine Learning as Search, Inductive Learning Hypothesis, Inductive Bias.
  • Classification/Interpolation Trees, (Generation, Evaluation, Interpretation), Generalizations/Extensions (Random Forests).
  • Data Mining Systems (Rapid Miner, Weka).
  • Case-Based Learning (k-NN, k-NN weighted distance), Case-Based Reasoning
  • Bayes Classifiers,
  • Support Vector Machines.
  • Clustering (divisive algorithms, hierarchical algorithms, density-based).
  • Association Rules.
  • Neural Networks (for classification or interpolation).
  • Combination of multiple models (Bagging, Boosting, Stacking).
  • Deep Learning, Deep Neural Networks.
  • Quality management in knowledge mining (evaluation of classification methods, model complexity (bias – variance), measures of interest of association rules, clustering validity, classification and interpolation evaluation metrics, ROC analysis).
  • Knowledge mining on the World Wide Web (opinion mining, sentiment analysis, fraud detection), shopping recommendations as an application of machine learning.
  • Data Mining: an overview of mining tasks, mining as an application of machine learning algorithms.

Recommended Bibliography

Useful